Re: [RFC] in-kernel rseq

[Mathieu Desnoyers]

On 2026-02-24 09:49, David Laight wrote:
> So all of the program has to use a single (per thread) 'rseq' structure?

Correct. One rseq TLS which contains a pointer to the current rseq_cs
descriptor (or NULL), and possibly many rseq_cs critical section
descriptors.

> And you better not try to use it in a signal handler.

The signal handler delivery handles the rseq abort on return to
userspace. The rseq critical section can be used within a signal
handler.

Thanks,

Mathieu

--
Mathieu Desnoyers
EfficiOS Inc.
https://www.efficios.com